Sensitive data

The Sensitive data tab shows which tables and columns hold personally identifiable information (PII) or protected health information (PHI), such as email addresses, phone numbers, or national IDs. Use it to see where sensitive data lives in each database.

The tab is available for PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, and MongoDB. To start scanning, turn on PII classification for the instance in the agent’s database monitoring settings. The tab shows results from the last seven days.

Read the results

The tiles across the top count the tables scanned, the tables with sensitive data, and the sensitive columns out of all columns scanned. They also name the classifier version that produced the results.

Each table in the list shows its highest sensitivity level, how many of its columns are sensitive, the categories found in it, and when it was last classified. Sensitivity levels run from Critical and High through Medium and Low to No PII. A column that couldn’t be classified is marked Unclassified.

Expand a table to see every column, with the sensitive columns listed first. Hover a column’s sensitivity to see its categories and how it was detected:

Detected by	Meaning
Dictionary	The column name matched a built-in list of sensitive column names.
Pattern	The column matched a pattern rule.
Model	The GLiNER-PII model detected sensitive data in the column, with a confidence score.
Cached	An earlier result for the same column was reused.
Override	Someone set the classification by hand. It takes precedence over automatic classification.

Caution

Only scanned tables appear in the list. A table that isn’t listed hasn’t been scanned yet, so it isn’t known to be free of sensitive data.

On an Oracle container database, tables live in the pluggable databases. Select a PDB in the Database menu to see its results.